The primary aim of the current application is to test a novel model of how the relative responsiveness of reward- and threat-related neural circuits interact with life stress to predict drug use in a large cohort of adolescents and young adults. The application also aims to identify genetic profiles representing variability in key neuromodulatory pathways that predict these interactions. To address these specific aims the application will collect measures of: 1) core behavioral processes that serve as risk factors for drug use and abuse, 2) brain circuits supporting threat and reward responsiveness as well as behavioral control, 3) the experience of stressful life events and 4) DNA sequence variation in 600 18-22 year old undergraduates. All measures will be collected using the existing extensive research infrastructure of the Duke Neurogenetics Study. Innovative analytic strategies for modeling individual differences will be applied to these data in the service of identifying behavioral, neurobiological, environmental and genetic mechanisms mediating variability in risk behaviors. The predictive utility of our findings for informing developmental trajectories of drug use, abuse, and dependence will be tested through federated studies. These studies are collecting overlapping measures in both community samples as well as longitudinal samples of patients and individuals at high risk (e.g., low SES, positive family history). Data federation wil also provide replication data sets and/or additional power for testing novel associations between neural, genetic, and environmental variation and the emergence of psychopathology. Furthermore, existing collaborations with investigators using animal models of human behavior and disease as well as those conducting epidemiological gene-environment interaction research will facilitate unique translational projects based on our findings. These collaborations will allo for identification of both detailed molecular mechanisms and broader clinical relevance. Collectively, the research proposed will reveal mechanisms and pathways that predict psychopathology with a focus on drug use and the transition to abuse, which represents a major thrust of current research supported by NIDA and is consistent with the Research Domain Criteria (RDoC) emphasis on classifying psychopathology based on dimensions of observable behavior and neurobiological measures.